xnli_xlm_r_only_fr
This model is a fine-tuned version of xlm-roberta-base on the xnli dataset. It achieves the following results on the evaluation set:
- Loss: 0.7620
- Accuracy: 0.7900
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.67 | 1.0 | 3068 | 0.5665 | 0.7667 |
| 0.5256 | 2.0 | 6136 | 0.5302 | 0.7932 |
| 0.4588 | 3.0 | 9204 | 0.5528 | 0.7884 |
| 0.4043 | 4.0 | 12272 | 0.5616 | 0.7867 |
| 0.3545 | 5.0 | 15340 | 0.5650 | 0.7984 |
| 0.3124 | 6.0 | 18408 | 0.6290 | 0.7912 |
| 0.275 | 7.0 | 21476 | 0.6616 | 0.7843 |
| 0.245 | 8.0 | 24544 | 0.6745 | 0.7932 |
| 0.2202 | 9.0 | 27612 | 0.7479 | 0.7900 |
| 0.2025 | 10.0 | 30680 | 0.7620 | 0.7900 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.6.1
- Tokenizers 0.13.1